1 / 16

Non-Parametric: K-NN

Non-Parametric: K-NN. Prof. A.L. Yuille Stat 231. Fall 2004. Duda, Hart and Stork: Chp 4.4 & 4.5. K-NN. K-NN. K-NN Examples. Non-parametric Classification. Non-Parametric Classification. Nearest Neighbour Decision Rule. Nearest Neighbour Decision Rule. Nearest Neighbour Decision Rule.

Télécharger la présentation

Non-Parametric: K-NN

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Non-Parametric: K-NN Prof. A.L. Yuille Stat 231. Fall 2004. Duda, Hart and Stork: Chp 4.4 & 4.5.

  2. K-NN.

  3. K-NN

  4. K-NN Examples.

  5. Non-parametric Classification.

  6. Non-Parametric Classification

  7. Nearest Neighbour Decision Rule

  8. Nearest Neighbour Decision Rule

  9. Nearest Neighbour Decision Rule

  10. Randomized Decisions.

  11. NN Error Analysis Here x* is the sample closest to point x. As the no. samples becomes large x* will be arbitrarily close to x.

  12. NN Error Analysis

  13. NN Error Analysis

  14. NN Error Analysis

  15. KNN Decision Rule

  16. NN Decision Rule • The nearest neighbour decision rule is very easy to use. • It can be effective in situations where other classification rules – e.g. linear separation – will not work. • The asymptotic result shows that the NN rule often approaches the performance of the (optimal) Bayes rule.

More Related